應用層級分析法建構共同基金選取模型之研究

Abstract

隨著經濟持續發展，國民所得提高，投資理財的觀念日益受到重視。金融改革與開放，使得各式各樣的投資工具如雨後春筍般出現。面對不斷推陳出新的金融產品，投資人常因投資金額、時間及專業知識等限制不得其門而入，而共同基金即以其可小額投資、專業人士操作管理、分散投資及流動性佳等優點，深受投資大眾的喜愛。截至九十三年二月份的統計，國內投信公司發行加計經核備海外基金總數超過國內股票上市、櫃公司家數。面對市場上琳瑯滿目的共同基金，投資人如何挑選出合適的標的，便成為基金投資時所須克服的一大難題。 衡諸目前市面上的書籍、雜誌及財務軟體，在提供投資人挑選共同基金的建議時，不是未考慮投資人個別的投資偏好，就是要求填入的篩選變數逾越投資人的財務知識，再者決策過程中常涉及主觀認定及黑箱作業。因此建立一套能滿足投資人個別偏好，同時兼顧客觀原則，且能具備數量化特性的共同基金選取模型，遂成為刻不容緩之事。 本研究提出了以層級分析法（Analytic Hierarchy Process，AHP）所建立的模型，用來解決選擇共同基金的複雜決策問題。透過模型嚴謹的架構，利用二階段的分析過程，協助個別投資人及財務顧問進行資產的配置及共同基金的選取。 第一階段資產配置方面，本模型利用個別投資人的問卷分析進行建議，所提出的是針對特定投資人量身訂作的客製化結果，能符合其個別的投資目標與限制。層級分析法模型的目的，並非僅只於建立一套在未來具有優異績效表現的投資組合，更重要的是要確保所建議的資產配置必須兼顧投資人的投資目標偏好。此即為本模型有別於其他基金評選模型之最大差異與貢獻。 第二階段排序選取共同基金方面，針對共同基金屬性的衡量，本研究除了傳統上風險與報酬因子外，更加入了其他的考量因素，包括三年總報酬率、夏普績效指標、費用比率、基金規模及經理人任期等，使得基金屬性衡量的範圍更為周詳且更貼近市場實際狀況，此為本模型的第二個貢獻。 最後本研究透過層級分析法模型，將各評估準則予以數量化，摒除以往只能主觀認定或黑箱作業的操作過程，建立一套客觀量化的基金評選機制。如此使得財務顧問在面對個別投資人時，能更合理、清晰且客觀地提出數據及資料進行基金的推薦，則為本模型的第三個貢獻。 本研究利用方便取樣的結果進行實證，證明以層級分析法為架構所建立的共同基金選取模型，除了具有操作簡便、富有彈性的優點外，更可以確保投資人及財務顧問在建構投資組合過程中決策的一致性。

Highly development of social and economic environment make financial planning a more important issue nowadays. Deregulation of financial and banking industry advance the booming of various investment tools. Great investment amount, constrain of time and professional background and knowledge become limitation for an individual investor while investing on these emergent financial products. Mutual funds provide small amount investment, professional management, diversification and liquidity for an individual investor. These features have helped stimulate the popularity of mutual funds to historic heights in the past decade. As of February 2004, the total number of domestic and offshore mutual funds is more than the IPO and OTC companies. The proliferation of mutual funds has made choosing the suitable funds a challenge to many investors. In response to the challenge described above, many books, magazines as well as financial softwares designed to assist investors in mutual fund selection. However, these sources range from inadequate to the useless. First, they do not consider the preference of individual investors. Second, few investors have sufficient financial knowledge to input the appropriate values for the screening variables. Finally, the overall process appears to be a “black box” and subjective judgment to the investor. It is urgent to develop a mutual fund selection model which presents not only quantitative character but also objective reality and can satisfy the preference of individual investors as well. In this research we propose and illustrate a structure approach to selecting mutual funds which framework is based on the analytic hierarchy process (AHP), in order to response the complicated decision problem of mutual fund selection. By using two stages approach, our model assists an individual investor and a financial advisor in determining an appropriate asset allocation and selecting mutual funds for a complete portfolio. The first stage is to determine the asset allocation. The asset allocations recommended by the AHP model are customized to fit the investment objectives and constraints of a particular investor. The AHP model is not intended to create a portfolio to provide superior future performance alone but to ensure that the asset allocation is consistent with the objectives and preferences of the investor. This is the biggest difference and the first contribution compared our model with others. The second stage is to rank and select mutual funds. In order to make the model more realistic, we evaluated on a total of five attributes of mutual funds in addition to risk and return, including Total Revenue in 3 Years, Sharpe Index, Expense Ratio, Fund Scale and Manager Tenure. This is the second contribution of our model. The AHP model quantifies each criterion of the hierarchy to construct an objective and quantitative mutual fund selection mechanism. The AHP approach helps financial advisors explain the fund selection process as well as recommend specific mutual funds to the investor in a more reasonable, clear and objective way. This is the third contribution of our research. We include sample mutual fund selection for several investors from convenience sampling. A mutual fund selection model that uses the AHP framework is flexible, is user friendly, and ensures consistency throughout the portfolio decision process.